Re: [PATCH net-next v4] net/sched: cap the accounted backlog before it can wrap

[Simon Horman <horms@kernel.org>]

On Wed, Oct 07, 2026 at 06:13:52AM -0400, Jamal Hadi Salim wrote:
> On Tue, Oct 6, 2026 at 11:30 AM Simon Horman <horms@kernel.org> wrote:
> >
> > On Fri, Oct 02, 2026 at 03:44:31AM -0400, Jamal Hadi Salim wrote:
> >
> > ...
> >
> > > diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
> >
> > ...
> >
> > > @@ -1712,18 +1738,24 @@ static struct cake_tin_data *cake_select_tin(struct Qdisc *sch,
> > >       return &qd->tins[tin];
> > >  }
> > >
> > > -static u32 cake_classify(struct Qdisc *sch, struct cake_tin_data **t,
> > > -                      struct sk_buff *skb, int flow_mode, int *qerr)
> > > +/* Run the optional external classifier. Returns true when the packet was
> > > + * consumed by a terminal TC action (reason in *qerr); else *flow/*host
> >
> > FYI:
> >
> > W=1 builds complain that the line above has a '/*' embedded in a comment.
> > This is likely why Sashiko NIPA didn't run for this series.
> 
> Thanks for catching this Simon. We don't run W=1 - already overwhelmed
> as it is with the flood. Do you think we should?

Yes.

It's important that patches don't add new warnings.
You can use some sort of diff mechanism to spot those amongst any noise.